Gene therapy holds enormous potential for a new era in human medicine. These methodologies will allow treatment for conditions that heretofore have not been addressable by standard medical practice. One area that is especially promising is the ability to genetically engineer a cell to cause that cell to express a product that has not previously been produced in that cell. Examples of uses of this technology include the insertion of a transgene encoding a novel therapeutic protein, insertion of a coding sequence encoding a protein that is lacking in the cell or in the individual, insertion of a wild type gene in a cell containing a mutated gene sequence, and insertion of a sequence that encodes a structural nucleic acid such as a microRNA or siRNA.
To meet the challenge of increasing global demand for food production, many effective approaches to improving agricultural productivity (e.g. enhanced yield or engineered pest resistance) rely on either mutation breeding or introduction of novel genes into the genomes of crop species by transformation. Both processes are inherently non-specific and relatively inefficient. For example, conventional plant transformation methods deliver exogenous DNA that integrates into the genome at random locations. The random nature of these methods makes it necessary to generate and screen hundreds of unique random-integration events per construct in order to identify and isolate transgenic lines with desirable attributes. Moreover, conventional transformation methods create several challenges for transgene evaluation including: (a) difficulty for predicting whether pleiotropic effects due to unintended genome disruption have occurred; and (b) difficulty for comparing the impact of different regulatory elements and transgene designs within a single transgene candidate, because such comparisons are complicated by random integration into the genome. As a result, conventional plant trait engineering is a laborious and cost intensive process with a low probability of success.
Precision gene modification overcomes the logistical challenges of conventional practices in plant systems, and as such has been a longstanding but elusive goal in both basic plant biology research and agricultural biotechnology. However, with the exception of “gene targeting” via positive-negative drug selection in rice or the use of pre-engineered restriction sites, targeted genome modification in all plant species, both model and crop, has until recently proven very difficult. Terada et al. (2002) Nat Biotechnol 20(10):1030; Terada et al. (2007) Plant Physiol 144(2):846; D'Halluin et al. (2008) Plant Biotechnology J. 6(1):93.
Transgene (or trait) stacking has great potential for production of plants, but has proven difficult. See, e.g., Halpin (2005) Plant Biotechnology Journal 3:141-155. In addition, polyploidy, where the organism has two or more duplicated (autoploidy) or related (alloploid) paired sets of chromosomes, occurs more often in plant species than in animals. For example, wheat has lines that are diploid (two sets of chromosomes), tetraploid (four sets of chromosomes) and hexaploid (six sets of chromosomes). In addition, many agriculturally important plants of the genus Brassica are also allotetraploids.
Transgenes can be delivered to a cell by a variety of ways, such that the transgene becomes integrated into the cell's own genome and is maintained there. In recent years, a strategy for transgene integration has been developed that uses cleavage with site-specific nucleases for targeted insertion into a chosen genomic locus (see, e.g., co-owned U.S. Pat. No. 7,888,121). Nucleases specific for targeted genes can be utilized such that the transgene construct is inserted by either homology directed repair (HDR) or by end capture during non-homologous end joining (NHEJ) driven processes. Targeted loci include “safe harbor” loci such as the AAVS1, HPRT and CCR5 genes in human cells, and Rosa26 in murine cells (see, e.g., U.S. Pat. Nos. 7,888,121; 7,972,854; 7,914,796; 7,951,925; 8,110,379; 8,409,861; 8,586,526; U.S. Patent Publication Nos. 2003/0232410; 2005/0208489; 2005/0026157; 2006/0063231; 2008/0159996; 2010/00218264; 2012/0017290; 2011/0265198; 2013/0137104; 2013/0122591; 2013/0177983 and 2013/0177960) and the Zp15 locus in plants (see U.S. Pat. No. 8,329,986). Nuclease-mediated integration offers the prospect of improved transgene expression, increased safety and expressional durability, as compared to classic integration approaches that rely on random integration of the transgene, since it allows exact transgene positioning for a minimal risk of gene silencing or activation of nearby oncogenes.
Genome engineering can also include the knocking out of genes in addition to insertion methods described above. In the absence of a donor nucleic acid, a cell with a cleaved genome will resort to the error prone NHEJ pathway to heal the break. This process often adds or deletes nucleotides during the repair process (“indels”) which may lead to the introduction of missense or non-sense mutations at the target site. For example, CCR5-specific zinc finger nucleases are being used in Phase I/II trials to create a non-functional CCR5 receptor, and thus prevent HIV infection (see U.S. Pat. No. 7,951,925).
Targeted nuclease-mediated genome cleavage at a desired location can be obtained by the use of an engineered nuclease. For example, a double-strand break (DSB) for can be created by a site-specific nuclease such as a zinc-finger nuclease (ZFN) or TAL effector domain nuclease (TALEN). See, for example, Urnov et al. (2010) Nature 435(7042):646-51; U.S. Pat. Nos. 8,586,526; 6,534,261; 6,599,692; 6,503,717; 6,689,558; 7,067,317; 7,262,054, the disclosures of which are incorporated by reference in their entireties for all purposes.
Another nuclease system involves the use of a so-called acquired immunity system found in bacteria and archaea known as the CRISPR/Cas system. CRISPR/Cas systems are found in 40% of bacteria and 90% of archaea and differ in the complexities of their systems. See, e.g., U.S. Pat. No. 8,697,359. The CRISPR loci (clustered regularly interspaced short palindromic repeat) is a region within the organism's genome where short segments of foreign DNA are integrated between short repeat palindromic sequences. These loci are transcribed and the RNA transcripts (“pre-crRNA”) are processed into short CRISPR RNAs (crRNAs). There are three types of CRISPR/Cas systems which all incorporate these RNAs and proteins known as “Cas” proteins (CRISPR associated). Types I and III both have Cas endonucleases that process the pre-crRNAs, that, when fully processed into crRNAs, assemble a multi-Cas protein complex that is capable of cleaving nucleic acids that are complementary to the crRNA.
In type II systems, crRNAs are produced using a different mechanism where a trans-activating RNA (tracrRNA) complementary to repeat sequences in the pre-crRNA, triggers processing by a double strand-specific RNase III in the presence of the Cas9 protein. Cas9 is then able to cleave a target DNA that is complementary to the mature crRNA however cleavage by Cas 9 is dependent both upon base-pairing between the crRNA and the target DNA, and on the presence of a short motif in the crRNA referred to as the PAM sequence (protospacer adjacent motif) (see Qi et al. (2013) Cell 152:1173). In addition, the tracrRNA must also be present as it base pairs with the crRNA at its 3′ end, and this association triggers Cas9 activity.
The Cas9 protein has at least two nuclease domains: one nuclease domain is similar to a HNH endonuclease, while the other resembles a Ruv endonuclease domain. The HNH-type domain appears to be responsible for cleaving the DNA strand that is complementary to the crRNA while the Ruv domain cleaves the non-complementary strand.
The requirement of the crRNA-tracrRNA complex can be avoided by use of an engineered “single-guide RNA” (sgRNA) that comprises the hairpin normally formed by the annealing of the crRNA and the tracrRNA (see Jinek et al. (2012) Science 337:816 and Cong et al. (2013) Sciencexpress/10.1126/science.1231143). In S. pyrogenes, the engineered tracrRNA:crRNA fusion, or the sgRNA, guides Cas9 to cleave the target DNA when a double strand RNA:DNA heterodimer forms between the Cas associated RNAs and the target DNA. This system comprising the Cas9 protein and an engineered sgRNA containing a PAM sequence has been used for RNA guided genome editing (see Ramalingam et al. (2013) Stem Cells and Development 22(4):595-610) and has been useful for zebrafish embryo genomic editing in vivo (see Hwang et al. (2013) Nature Biotechnology 31(3):227) with editing efficiencies similar to ZFNs and TALENs.
Thus, there remains a need for systems for genomic editing, including for treatment and/or prevention of diseases and for agricultural uses.